CoCorrRec applies test-time training to on-device sequential recommendation, with a cloud-side correction network, and reports accuracy gains over RNN and Transformer baselines on three datasets.
DUET: A tuning-free device-cloud collaborative parame- ters generation framework for efficient device model gen- eralization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Device-Cloud Collaborative Correction for On-Device Recommendation
CoCorrRec applies test-time training to on-device sequential recommendation, with a cloud-side correction network, and reports accuracy gains over RNN and Transformer baselines on three datasets.